When Homo sapiens began to develop technology, one of the first inventions they devised was alloys. We all know steel made of iron and carbon or bronze made of copper and tin. The idea is to use a first compound and add a second to reinforce or drastically change some of the original properties of the first and thus develop new technological applications.
Much more recently, scientists began to wonder what effect it would have to randomly change atoms of a metal, such as copper, and replace them with magnetic atoms, such as iron or manganese. Thus began the production and study of the properties of materials called spin glasses. They are so named because these materials behave in a very similar way to the stained glass windows in cathedrals, since these glasses have enormous relaxation times – the atoms in the stained glass windows have not yet reached equilibrium – and are made up of metals with electrons ( with its spin) and magnetic atoms.
As we have randomly substituted metal atoms, the electrons in the metal live in a very messy environment and this produces a special kind of interaction between these electrons and the magnetic impurities introduced. In turn, this disorder also produces an effect of frustration , which makes it very difficult for the elements of the spin glass to come together and “work” in unison. It happens exactly the same as in a human group: if A is a friend of B and an enemy of C, will B and C be friends or enemies? The combined effect of disorder and frustration make these materials very interesting examples of complex systems, in which a system is much more than the sum of its parts.
Amazing behavior of spin glasses
The spin glasses have, in addition to the previously described behaviors, very attractive properties. Thus, and from the formal point of view, Giorgio Parisi, theoretical physicist at the University of Rome La Sapienza, found, by solving a simpler version of the equations that describe them, that these glasses should present very surprising properties when they were cooled to very low temperatures (about -250 ºC): a very large -infinite- number of phases that would be organized in a way very similar to biological species in a taxonomic tree. Let us remember that matter can appear in a liquid, gas or solid phase, and that magnets can have paramagnetic phases (without magnetization) or ferromagnetic (with magnetization) depending on their temperature.
This solution found by Parisi describes quite well, although there are still scientists who disagree, the real spin glasses. To verify it, a huge effort has been necessary, fundamentally computational, which has made necessary the development of new simulation algorithms and also the construction of new dedicated computers, that is, computers that are extraordinarily efficient in simulating a class of systems, but are very ineffective simulating the rest. We can cite the Janus family of computers developed within a Spanish-Italian collaboration (Extremadura-Madrid-Zaragoza-Ferrara-Roma). These computers are based on electronic devices called FPGAs (many domestic appliances have one) that allow a simulation time of around several months to reproduce almost a second of an experiment, which has allowed to verify the theoretical results with real experiments.
Finally, it should be noted that although these spin glasses are not currently used in the design of technological devices, many of the theoretical results found are being applied with great success in different fields of science and technology (fundamentally in solving complex problems and optimization ) and they are also being very relevant in the study of normal glass (those of cathedrals), thus closing the circle, in some way.
Juan Jesús Ruiz-Lorenzo is a tenured professor of Theoretical Physics at the University of Extremadura and a member of the Statistical Physics Research Group (SPhinX)
